This Presentation talks about:
-Introduction to AI
-Introduction to Chatbots
-Why do we need Conversational Agents
-Use Cases
-Watson Assistant
-Building Blocks
-Analytics
3. • Introduction to AI
• Introduction to Chatbots
o Why do we need Conversational Agents
o Use Cases
• Watson Assistant
o Building Blocks
o Analytics
• Demo
Get started at: https://ibm.biz/assistantchatbot
Agenda
5. Artificial Intelligence (AI) refers to
the ability of computer systems to
perform tasks commonly
with intelligent beings.
These processes include learning,
reasoning, and self-correction.
7. What is a Chatbot?
An Artificial Intelligence tool that can
human conversation, with the objective of
providing some value to the end user.
Using chatbots you can get information,
pay bills, provide feedback, the
are endless...
10. IBM Watson Assistant
10
Create an application that understands natural-language and responds to customers in human-
like conversation –in multiple languages.
Seamlessly connect to messaging channels, web environments and social networks to make
scaling easy.
Easily configure a workspace and develop your application to suit your needs.
• Expose your conversation as APIs
• Supports multiple languages (including Arabic)
• Can understand context
• Doesn’t require coding skills to develop*
* However requires coding to integrate
12. Components of Assistant
Intents: An Intent is the purpose behind a statement.
Entities: Entities are like the nouns in a sentence.They are
used to clarify a user’s intent
Dialog: The Dialog is the flow of the conversation.
13. Components of Assistant
13
INTENTS
Identifies the purpose behind the user’s query
1. Where is the office ?
2. How can I get to the swimming pool?
3. Is there a bank here?
4. I want to book a room for 2.
Intent 1, 2, 3: #place
Intent 4: #booking
16. Components of Assistant
16
ENTITIES
Identifies the object of the user’s query
1. Where is the office?
2. How can I get to swimming pool?
3. Is there a bank here?
4. I want to book a room for 2.
Entity 1: @location: office
Entity 2: @location: swimming pool
Entity 3: @location: bank
Entity 4: @room: room, @number: 2
18. 18
DIALOG
Defines the flow of the conversation.
Used to identify what conditions trigger what
response. Built using a combination of intents,
entities, and context variables.
• Conditional
• Folders – Group similar content, handle digression
• Prebuilt Content
Avoid having a tree more than 20 nodes deep.
Components of Assistant
19. Analytics
• Metrics Overview
• Improve your skill
• Empower your skill to learn automatically
• Defining what's irrelevant
20. Demo
Sign-up or Sign-in at IBM Cloud: https://ibm.biz/assistantchatbot
21. • Artificial Intelligence (AI) refers to the ability of computer systems to perform tasks commonly
associated with intelligent beings.
• A Chatbot is an Artificial Intelligence tool that can imitate human conversation, with the
objective of providing some value to the end user.
• Why do we need chatbots and why are they important for businesses.
• Siri and Bixbi as real-life use cases for chatbots that everyone is using for different purposes.
• Watson Assistant service on IBM Cloud, its building blocks (Intents, Entities and Dialogs). We
have also covered the analytics part of the service.
• Lastly, we have concluded the presentation with a demo for deploying a Watson Assistant
service to IBM Cloud.
Wrap Up
So what is AI ?
AI refers to the ability of a computer system to act like a human being by performing tasks that you would normally associate with humans, so tasks like learning communicating, self correction etc.
A common misconception about AI is that, it is supposed to replace humans. However, that’s not exactly accurate. AI as I see it is supposed to augment or enhance our abilities as humans, to allow us to solve even more complex problems.
So building on that point, some of the applications of AI include sifting large amounts of data and deriving insights from that data which can then help us make better decisions, these decisions can include predicting weather patterns, predicting the stock market trends etc. AI can also be used to create tools like chatbots which can interact with 100’s of people at the same time. So so you all these applications of AI, they are used to augment human abilities. And you will see and example of one such use-case of AI in today’s demo
So now that you have an overview of what AI is, let me tell you alittle bit more about chatbots
let me explain to you guys what a chatbot is, so a chatbot is an AI tool that can imitate human conversation. So what happens is when you are chatting with a chatbot and you send your message, the chatbot uses AI to identify the user’s intent behind that message, it tried to determine what is user is looking for or what the user wants to know. And then once the chatbot has figured that out, it provides the most appropriate response to the users query.
Chatbots have many applications, chatbots can be integrated with websites to make it easier for the users to browse through the website, they can also be used in applications to answers some questions that are very frequently asked by the users, or to get feedback from the users etc. so you see there are a lot of applications and uses that chatbots can have.
So let me actually tell you about some of the commons applications of chatbots that you guys might have interacted with.
For any industry type, if you take account the user base, there are some basic questions that users generally ask. And for these specific question, same specific answers are required to be answered every time the customer makes the similar query. Well, for such purposes, chatbot is the best engaging way to answer these common questions.
Many businesses use chatbots to make sure their internal business processes are more efficient. Chatbots can manage calendars, hunt down documentation, and even make it easier to manage assets, such as social media. HR departments are also using chatbots to handle internal requests. Managers can use bots to send out messages and oversee the team.
One of the great advantages of chatbot is, it requires less development cost in comparison to application development. Well, making an investment into a quality product like chatbot is worth as it can offer a better experience to your customers.
The best example of a chatbot is SIRI, which is present in the iphone. For those of you who have used an iphone would know that you can se siri to do various tasks on your iphone. You can use it to make phone calls, set reminders, set alarms, and so on. And it is as simple as having a conversation, you can ask siri to make a phone call to someone in your contacts and it will do that.
Similar to Siri, for Samsung users. Samsung phone also have their own chabot which is called bixby.
So these are some of the examples of chatbots that you guys might have interacted it.
IBM Watson™ Assistant is used to build your own branded live chatbot into any device, application, or channel. Your chatbot, which is also known as an assistant, connects to the customer engagement resources you already use to deliver an engaging, unified problem-solving experience to your customers.
With Watson conversation, you can add a built-in channel integration to deploy the configured assistant directly to a social media or messaging channel, such as Facebook Messenger, Slack, Voice Agent ..
Build your own client application as the user interface for the assistant. Or add the built-in web chat integration to your company website. From the web chat you can transfer customers who ask to speak to someone to your existing service desk personnel.Supports up to 12 languages
And now lets look into the major components of a chabot.
The three main components of a chatbot are intents, entities and dialogs. Its as simple as that, you put all these three components together and you have a chatbot. So I want you all to remember these three components because we will be using these in our demo. And not lets look into what these components actually are and what is their function in a chatbot.
Intent, as the name suggests is the purpose of the users statement, it is basically what the user intends when they communicate something to the chatbot. So if I say to a chabot, set reminder, my intention behind that statement is that I want to set a reminder. Simple right ? But its very important for the Chabot to categorize your statements according to the right intents because it is only then that the chatbot will be able to guide you correctly or provide you with accurate information etc.
Entities, are used to further clarify the users intent. So e.g. in the previous example like I mentioned if I say to the Chabot “set a reminder” my intention is to set a reminder. But the chatbot needs more information about my intent in order to help me appropriately. E.g. the chatbot needs to know what time the reminder is for and what date the reminder is for. So all this additional information needed to further explain the users intent, comes under entities.
Lastly, the dialog. Dialog is the flow of the conversation. It is basically how the chatbot will actually respond to the users queries. And for that we use the intents and entities to. So now building on our earlier example: once I say to a chatbot, “set a reminder” it will categorize my intent is to set a reminder and it will ask me a question such as “What is the time and date of the reminder” etc. So this whole flow of the conversation…me asking the chatbot to set a reminder and then the chatbot asking me additional questions is referred to as the dialog.
A vey important aspect of Watson assistant for all businesses is the analytics. With the analytics, you can gain insights on how your assistant is
The Overview page provides a summary of the interactions between users and your assistant. You can view the amount of traffic for a given time period, as well as the intents and entities that were recognized most often in user conversations.
The Analytics page of Watson Assistant provides a history of conversations between users and a deployed assistant. You can use this history to improve how your assistants understand and respond to user requests.
When customers interact with your assistant, they often make choices. If your underlying dialog skill pays attention, it can learn from these user decisions over time.
Teach your dialog skill to recognize when a user asks about topics that it is not designed to answer. And training your assistant on how to ignore specific subjects.